[FIX] queue_job: max retry #622
Conversation
When job fails because of concurrent update error, it does not respect max retries set by the job. Problem is ``perform`` method logic that handles re-try is never called, because ``runjob`` in controller that triggers jobs, catches expected exception and silences it. Though it is done to not pollute logs. So for now, adding extra check before job is run, to make sure max retries are handled if it reached it.
|
Hi @guewen, |
|
There hasn't been any activity on this pull request in the past 4 months, so it has been marked as stale and it will be closed automatically if no further activity occurs in the next 30 days. |
github-actions
bot
added
the
stale
|
@guewen can you check this? |
amh-mw
left a comment
amh-mw
left a comment
There was a problem hiding this comment.
Seems reasonable, but broad enough in scope that it should be covered by unit tests?
github-actions
bot
removed
the
stale
|
There hasn't been any activity on this pull request in the past 4 months, so it has been marked as stale and it will be closed automatically if no further activity occurs in the next 30 days. |
github-actions
bot
added
the
stale
| job.store() | ||
| env.cr.commit() | ||
|
|
||
| # FIXME: this exception never triggers up, so it never reaches |
There was a problem hiding this comment.
Don't remember exact scenario how this would happen, but it would just keep running and retrying way past max retries, if let say concurrent update error would occur over and over again.
|
I found this issue was happening in our customer environment. @simahawk Can you please reopen the PR? |
|
Hi @guewen, |
github-actions
bot
removed
the
stale
6 participants
When job fails because of concurrent update error, it does not respect max retries set by the job. Problem is that
performmethod logic that handles re-try is never called, becauserunjobin controller that triggers jobs, catches expected exception and silences it. Though it is done to not pollute logs.So for now, adding extra check before job is run, to make sure max retries are handled if it reached it.
Some context:
It looks like code that supposed to handle max retries, is never called. But I am not sure what would be the right way to trigger up exception as there is some logic in here
queue/queue_job/controllers/main.py
Line 125 in e2c6bab
Not having max retries can be very problematic if your jobs can have many concurrent updates. Had some issue where somehow same job record (yes job record itself, not some other records, job would update) was being updated by two job runners at the same time and it would always fail and re-try. It had over 400 re-tries. And the only way to stop it, was to restart odoo.
For example, without this fix we can end up in situation like this: